Why it matters
Without it, misalignment between WSI and field reality can cause missed structural features, delayed reporting, and repeated surveys, increasing risk to consent conditions and archive deposition.
From design to safety
Design
AI plans recording and phasing
AI reads the written scheme of investigation, drawings, and expected elevations to propose TLS positions, camera coverage, and control point layout for single context recording, including baseline survey overlaps and redundancy for critical elevations.
- terrestrial laser scanner
- digital camera
- prism and reflector targets
- total station
Build
AI guides field capture workflow
AI tracks progress against the WSI, flags gaps in elevations, levels, and photo sets, directs re-surveys to maintain complete context sheets, and logs re-survey reason codes for traceability.
- terrestrial laser scanner
- digital camera
- prism and reflector targets
- total station
Operation
AI curates ongoing models and reports
AI generates 3D models, strip maps, and context sheets from new scans and photos, triggers watching briefs when features align with Harris matrix, and exports staged reports for HER submissions.
- terrestrial laser scanner
- digital camera
- prism and reflector targets
- total station
Maintenance
AI preserves data integrity
AI validates calibration records, updates photogrammetry tie-points, checks archive deposition readiness, and flags outliers for post-excavation assessment, maintaining versioned back-ups and audit trails.
- terrestrial laser scanner
- digital camera
- prism and reflector targets
- total station
Safety
AI supports site safety through recording integrity
AI cross-references elevations with trench edges and buried services to reduce risk during ongoing excavation and ensure compliance with planning conditions, while logging safety-critical changes in the HER file.
- terrestrial laser scanner
- digital camera
- prism and reflector targets
- total station
Part by part
- 1
terrestrial laser scanner
- Watch for
- Scan gaps from occlusions or reflective surfaces that reduce point density in elevations.
- AI action
- AI detects coverage gaps and schedules additional scans or panoramas.
- 2
digital camera
- Watch for
- Blurred or poorly exposed frames that degrade context sheets and plans.
- AI action
- AI prompts retakes and adjusts exposure settings for accurate photogrammetry.
- 3
prism and reflector targets
- Watch for
- Misplaced or damaged targets that distort scale and tie-points.
- AI action
- AI verifies target geometry and recalibrates scale for models.
- 4
total station
- Watch for
- Instrument drift or mispointing affecting elevations and coordinates.
- AI action
- AI runs internal checks and issues calibration or re-pointing commands.
- 5
control points array
- Watch for
- Point coordinates drifting due to base resection errors or misidentified ties.
- AI action
- AI cross-validates with TLS and recalculates local grid before export.
- 6
scale bar
- Watch for
- Scale bar misalignment or distortion affecting contextual scales.
- AI action
- AI flags scale anomalies and prompts re-measurement with known references.
Worked example
Catching a misaligned elevation in a 1.2 ha site
Illustrative scenario with sample data.
A development-led trenching program requires accurate elevations for context mapping and archive deposition.
Signal in
siteArea: 1.2; periodsFound: 4; wsis: present; TLS: 2 units, camera: 4; controlPoints: 12; elevations: 1.5m–3.2m; scale: 1:50; targets: 24; referenceGrid: A0
System out
AI flags a 0.4m elevation mismatch between two trenches; generates corrective scan plan and updated context sheets; logs to archive; notes potential impact on Harris matrix ordering.
Human sign-off: Project manager reviews the AI plan, authorizes the re-survey, and approves updated context sheets for HER submission; heritage consultant signs off on revised phasing and archive metadata.
How to roll it out
Phase 1
Weeks 1 to 4: Baseline
Establish core recording hardware, initial WSI alignment, and baseline AI calibration against known elevations.
Phase 2
Weeks 5 to 12: Expansion
Scale up TLS and photogrammetry coverage, integrate with context sheets and Harris matrix, start automated reporting.
Phase 3
Weeks 13+: Maturation
Automated archive deposition workflows, ongoing calibration checks, and safety-linked recording thresholds.
Where people stay in charge
- Decisions about legal compliance, HER submissions, or consent conditions must be signed by the project manager or heritage consultant.
- Human oversight is mandatory for marked discrepancies in elevations, scale, or feature interpretation before reporting to planning authorities.
- Events that alter scope of works or WSI require formal re-approval from the client’s planning archaeologist.
Questions owners and crews ask
What constitutes the core AI outputs for standing building recording on excavation sites?
AI outputs include updated context sheets, photogrammetric models, elevation data, and phasing plans, with change logs and a complete running archive log for audit.
How does AI handle calibration and accuracy across multiple devices?
AI compares tie-points and control points across scans, flags drift, and prompts recalibration by the survey team to maintain consistent accuracy.
Who approves the generated models and reports?
The project manager and heritage consultant jointly review and sign off on models, context sheets, and archive deposition readiness.
What happens if a data gap is detected late in the build phase?
AI generates a re-survey plan, notifies the supervisor, and schedules additional captures to close the gap within the WSI.